In recent years, in the fields of thermoelectric conversion elements, field-effect transistors, sensors, integrated circuits, rectifying elements, photovoltaic cells, catalysts, electroluminescence, and the like, attention has been drawn to use of nanomaterials for making a flexible element or a small and light-weight element.
In the above fields, it is typically preferable to use a bipolar element which includes both a material exhibiting p-type conductivity and a material exhibiting n-type conductivity. For example, thermoelectric conversion elements are elements used for thermoelectric generation, in which power is generated by utilizing potential difference which occurs in a substance due to temperature difference. According to a thermoelectric conversion element including only one of a thermoelectric conversion material exhibiting p-type conductivity and a thermoelectric conversion material exhibiting n-type conductivity, power generation efficiency is poor. This is because heat is lost through a high-temperature-side terminal. FIG. 1 is a view schematically illustrating a bipolar thermoelectric conversion element which employs both a thermoelectric conversion material having n-type conductivity (n-type material) and a thermoelectric conversion material having p-type conductivity (p-type material). In a case where such a bipolar thermoelectric conversion element is used, power can be efficiently generated by connecting the thermoelectric conversion material having n-type conductivity and the thermoelectric conversion material having p-type conductivity in series.
Patent Literature 1 and Non-Patent Literature 1 each disclose a thermoelectric conversion material containing a carbon nanotube. The carbon nanotube utilized in technologies disclosed in Patent Literature 1 and Non-Patent Literature 1 is mainly a nanomaterial having p-type conductivity. As in the cases disclosed in Patent Literature 1 and Non-Patent Literature 1, nanomaterials often exhibit p-type conductivity. Accordingly, there is a demand for a technique for converting a nanomaterial exhibiting p-type conductivity into a nanomaterial exhibiting n-type conductivity. Note that a polarity of a nanomaterial (that is, whether a nanomaterial exhibits p-type conductivity or n-type conductivity) can be determined depending on whether a Seebeck coefficient is positive or negative. In other words, the technique for converting a polarity of a nanomaterial is a technique for changing a Seebeck coefficient.
For example, studies are made on conversion of a carbon nanotube exhibiting p-type conductivity into a carbon nanotube exhibiting n-type conductivity. So far, it has been reported that nitrogen atom exchange, alkali metal doping, or the like allows for conversion of a carbon nanotube having p-type conductivity into a carbon nanotube having n-type conductivity (for example, see Non-Patent Literatures 2 and 3).
Further, Non-Patent Literatures 4 and 5 each disclose that doping of a carbon nanotube with polyethylenimine allows for conversion of a carbon nanotube having p-type conductivity into a carbon nanotube having n-type conductivity. In addition, it has been reported that benzylviologen, ammonium and nicotinamide each can be also used as a dopant for converting a carbon nanotube having p-type conductivity into a carbon nanotube having n-type conductivity (see Non-Patent Literature 6, and Patent Literatures 2 and 3).
Furthermore, the inventors of the present invention have found that: several dopants are each capable of converting a single walled carbon nanotube having p-type conductivity into a single walled carbon nanotube having n-type conductivity; and each of these dopants has a HOMO level in a specific range (see Non-Patent Literature 7).